The long-term goal of this research is to understand how peptidergic neurons contribute to information processing in the primate retina. Retinal peptides were previously thought to act primarily as neuromodulators, and the localizations of immunoreactive peptides in wide-field amacrine, associational and interplexiform cells that made the majority of their contacts with amacrine cells seemed to support this hypothesis. During the first grant period, however, some peptidergic amacrine were found to make extensive contacts with bipolar and ganglion cells. The most striking exception to the earlier generalizations about peptide function was the localization of a peptide to a type of bipolar cell. These results in the first grant period suggested that peptidergic cells are not only interacting with the most direct pathway for visual information; they also appear to be used as neurotransmitters by the cells that comprise that pathway. In the experiments proposed in this application for renewal, the emphasis will shift from the peptides, themselves, to the functions of the cells that contain them. Light microscopic double label experiments will be designed to identify the full complement of chemical messengers in each type of cell. Based on those results, electron microscopic double label experiments will be designed to identify the cells contacted by the peptidergic neurons, and other neurons in the circuits will be identified by serial reconstruction of electron micrographs. Studies with cryofixed retina will determine whether peptides are stored in and, presumably, released from synaptic vesicles. The light microscopic techniques developed for the macaque monkey retina will also be applied to the human retina, and the results would provide the basis for studies of peptidergic neurons in diseased eyes. Recent studies showing changes in retinal peptide levels after experimentally induced diabetes suggest that these neurons may be selectively affected by retinal diseases. The proposed experiments with peptidergic bipolar cells are particularly likely to have clinical relevance since they appear to contact the short wavelength cones. The short wavelength system is particularly vulnerable in the early stages of a number of retinal diseases, and the proposed basic research on the synaptic interactions in that pathway might facilitate the interpretation of the deficits.